The Salad Bowl

Search Result, including tone of the article and text to speech news
Landing page

Inspiration

We were inspired by the lack of direct access to news that interest us. We don't want to spend a lot of time browsing multiple news websites to get summarized news.

What it does

The application crawls over 5000 websites and summarizes the most recent and relevant news from a specified keyword in a couple sentences and categorize them using the tones of the articles. The news are displayed for quick reading and only the most important sentences from the original news article are chosen in the summary. No more duplicate reading! The application can even read the news in audio form.

How we built it

We used Node.js and Express as our back-end to map api calls. We used React for the front-end. Using a natural langual processing algorithm, we implemented a text summarizer that extracts the most important sentences in a news article. Using IBM Watson, we extracted the tones expressed to categorize the news articles. Watson was also used to get an audio file of the news.

Challenges we ran into

The main challenge we ran into was coming up with a project definition. We had a general idea but we couldn't figure what we wanted to do with it.

Accomplishments that we're proud of

Having a working product in a short amount of time. The final product is polished and we implemented features from sponsor's API.

What we learned

Marketing should be even more important in the development of our product to avoid adding features that add no value to the user.

What's next for The Salad Bowl

We see Salad Bowl as the future platform to access news fast and accurately. To achieve this we plan on adding customized search for every user supported by data analytics.